Study on the Use and Adaptation of Bottleneck Features for Robust Speech Recognition of Nonlinearly Distorted Speech.
Jirí MálekPetr CervaLadislav SepsJan NouzaPublished in: SIGMAP (2016)
Keyphrases
- speech recognition
- speech recognition systems
- noisy environments
- speech synthesis
- speech signal
- automatic speech recognition
- hidden markov models
- pattern recognition
- language model
- speaker adaptation
- speech processing
- cepstral coefficients
- speech recognition technology
- handwriting recognition
- speech recognizer
- speech retrieval
- speaker identification
- feature extraction
- speaker recognition
- classification accuracy
- speech recognition errors
- speech recognizers
- mel frequency cepstral coefficients
- recognition engine
- keyword spotting
- word error rate
- speaker independent
- feature set
- speaker dependent
- noisy speech
- machine learning
- isolated word